Probabilistic neural network based olfactory classification for household burning in early fire detection application

Allan M. Andrew, Kamarulzaman Kamarudin, S.M Mamduh, Ali Yeon Md. Shakaff, Ammar Zakaria, Abdul H. Adom, David Lorater Ndzi, S. Ragunathan

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Determination of burning smell is important because it can help in early fire detection and prevention. In this paper, a household burning smell classification system for early fire detection application has been proposed using Probabilistic Neural Network (PNN) and PCA analysis. The experiments were performed on recorded smell samples from combustion of ten different commonly available household, including candle, joss sticks, air freshener, mosquito coil, newspaper, card board, plastic materials, Styrofoam and wood. All the experiments were done in a test chamber with humidity and temperature sensors. Portable Electronic Nose (PEN3) from Airsense Analytics is used as the measurement device. The smell source is placed 0.3m from the PEN3 and the time-series signal is measured for two minutes. The odour metrics is modelled using Probabilistic Neural Network. It is found that the average classification accuracy for the model is 99.62%.
Original languageEnglish
Title of host publicationIEEE Conference on Open Systems (ICOS 2013), 2013
PublisherIEEE
Pages221-225
Number of pages5
ISBN (Electronic)978-1-4799-0285-9, 978-1-4799-3152-1
ISBN (Print)9781479902866
DOIs
Publication statusPublished - 10 Feb 2014
Externally publishedYes

Fingerprint

Fires
Neural networks
Humidity sensors
Odors
Temperature sensors
Time series
Wood
Experiments
Plastics
Air
Electronic nose

Keywords

  • olfactory
  • neural network
  • classification
  • time series signal
  • fire detection

Cite this

Andrew, A. M., Kamarudin, K., Mamduh, S. M., Shakaff, A. Y. M., Zakaria, A., Adom, A. H., ... Ragunathan, S. (2014). Probabilistic neural network based olfactory classification for household burning in early fire detection application. In IEEE Conference on Open Systems (ICOS 2013), 2013 (pp. 221-225). IEEE. https://doi.org/10.1109/ICOS.2013.6735078
Andrew, Allan M. ; Kamarudin, Kamarulzaman ; Mamduh, S.M ; Shakaff, Ali Yeon Md. ; Zakaria, Ammar ; Adom, Abdul H. ; Ndzi, David Lorater ; Ragunathan, S. / Probabilistic neural network based olfactory classification for household burning in early fire detection application. IEEE Conference on Open Systems (ICOS 2013), 2013 . IEEE, 2014. pp. 221-225
@inproceedings{135b88ff690f481b8ccda0f36a6ef5eb,
title = "Probabilistic neural network based olfactory classification for household burning in early fire detection application",
abstract = "Determination of burning smell is important because it can help in early fire detection and prevention. In this paper, a household burning smell classification system for early fire detection application has been proposed using Probabilistic Neural Network (PNN) and PCA analysis. The experiments were performed on recorded smell samples from combustion of ten different commonly available household, including candle, joss sticks, air freshener, mosquito coil, newspaper, card board, plastic materials, Styrofoam and wood. All the experiments were done in a test chamber with humidity and temperature sensors. Portable Electronic Nose (PEN3) from Airsense Analytics is used as the measurement device. The smell source is placed 0.3m from the PEN3 and the time-series signal is measured for two minutes. The odour metrics is modelled using Probabilistic Neural Network. It is found that the average classification accuracy for the model is 99.62{\%}.",
keywords = "olfactory, neural network, classification, time series signal, fire detection",
author = "Andrew, {Allan M.} and Kamarulzaman Kamarudin and S.M Mamduh and Shakaff, {Ali Yeon Md.} and Ammar Zakaria and Adom, {Abdul H.} and Ndzi, {David Lorater} and S. Ragunathan",
year = "2014",
month = "2",
day = "10",
doi = "10.1109/ICOS.2013.6735078",
language = "English",
isbn = "9781479902866",
pages = "221--225",
booktitle = "IEEE Conference on Open Systems (ICOS 2013), 2013",
publisher = "IEEE",
address = "United States",

}

Andrew, AM, Kamarudin, K, Mamduh, SM, Shakaff, AYM, Zakaria, A, Adom, AH, Ndzi, DL & Ragunathan, S 2014, Probabilistic neural network based olfactory classification for household burning in early fire detection application. in IEEE Conference on Open Systems (ICOS 2013), 2013 . IEEE, pp. 221-225. https://doi.org/10.1109/ICOS.2013.6735078

Probabilistic neural network based olfactory classification for household burning in early fire detection application. / Andrew, Allan M.; Kamarudin, Kamarulzaman; Mamduh, S.M; Shakaff, Ali Yeon Md.; Zakaria, Ammar; Adom, Abdul H.; Ndzi, David Lorater; Ragunathan, S.

IEEE Conference on Open Systems (ICOS 2013), 2013 . IEEE, 2014. p. 221-225.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Probabilistic neural network based olfactory classification for household burning in early fire detection application

AU - Andrew, Allan M.

AU - Kamarudin, Kamarulzaman

AU - Mamduh, S.M

AU - Shakaff, Ali Yeon Md.

AU - Zakaria, Ammar

AU - Adom, Abdul H.

AU - Ndzi, David Lorater

AU - Ragunathan, S.

PY - 2014/2/10

Y1 - 2014/2/10

N2 - Determination of burning smell is important because it can help in early fire detection and prevention. In this paper, a household burning smell classification system for early fire detection application has been proposed using Probabilistic Neural Network (PNN) and PCA analysis. The experiments were performed on recorded smell samples from combustion of ten different commonly available household, including candle, joss sticks, air freshener, mosquito coil, newspaper, card board, plastic materials, Styrofoam and wood. All the experiments were done in a test chamber with humidity and temperature sensors. Portable Electronic Nose (PEN3) from Airsense Analytics is used as the measurement device. The smell source is placed 0.3m from the PEN3 and the time-series signal is measured for two minutes. The odour metrics is modelled using Probabilistic Neural Network. It is found that the average classification accuracy for the model is 99.62%.

AB - Determination of burning smell is important because it can help in early fire detection and prevention. In this paper, a household burning smell classification system for early fire detection application has been proposed using Probabilistic Neural Network (PNN) and PCA analysis. The experiments were performed on recorded smell samples from combustion of ten different commonly available household, including candle, joss sticks, air freshener, mosquito coil, newspaper, card board, plastic materials, Styrofoam and wood. All the experiments were done in a test chamber with humidity and temperature sensors. Portable Electronic Nose (PEN3) from Airsense Analytics is used as the measurement device. The smell source is placed 0.3m from the PEN3 and the time-series signal is measured for two minutes. The odour metrics is modelled using Probabilistic Neural Network. It is found that the average classification accuracy for the model is 99.62%.

KW - olfactory

KW - neural network

KW - classification

KW - time series signal

KW - fire detection

U2 - 10.1109/ICOS.2013.6735078

DO - 10.1109/ICOS.2013.6735078

M3 - Conference contribution

SN - 9781479902866

SP - 221

EP - 225

BT - IEEE Conference on Open Systems (ICOS 2013), 2013

PB - IEEE

ER -

Andrew AM, Kamarudin K, Mamduh SM, Shakaff AYM, Zakaria A, Adom AH et al. Probabilistic neural network based olfactory classification for household burning in early fire detection application. In IEEE Conference on Open Systems (ICOS 2013), 2013 . IEEE. 2014. p. 221-225 https://doi.org/10.1109/ICOS.2013.6735078